The Change Paradox - Why Organizations Fight the Change They Say They Want

Sometimes organizations do something that makes very little sense from the outside.

A department has struggled for years. Performance is poor. Turnover is high. Trust has deteriorated. Problems that should have been addressed long ago have become embedded in the culture.

And leadership knows it.

They may spend months, or even more than a year, trying to recruit someone capable of fixing it. During interviews they talk openly about the need for change. They want stronger accountability, healthier culture, better productivity, improved retention, and better performance indicators.

Eventually, they hire a leader precisely because things need to change—and give that person the responsibility for making it happen.

Then the new leader starts changing things. And the organization begins trying to stop them.

I have watched versions of this happen in healthcare, education, business, churches, and nonprofits. The details change, but the pattern is remarkably consistent.

Organizations frequently want different results without changing the system that produces the current results.

This is one form of the change paradox—wanting different results without changing the system that produces them.

Wanting Change and Accepting Change Are Different Decisions

When executives say they want change, they are usually sincere about the outcome they desire. They really do want better performance and lower turnover. They really do want stronger culture. They really do want people held accountable.

The problem appears when those abstract goals take on names, consequences, and social or cultural costs.

Accountability may mean confronting a popular employee who has been creating problems for years. Improving culture may mean reducing the influence of someone with long-standing relationships throughout the organization. Raising standards may cause some employees to resign. Now protecting high performers may require disciplining people who have learned that consequences are negotiable.

Changing expectations may temporarily increase complaints and replacing poor performers may initially make staffing metrics worse rather than better.

Suddenly leadership discovers that the change they wanted has a price. And that is where many turnarounds fail. But often, organizations demonstrate that they do not want change more than they are willing to pay for that change.

Bad Systems Can Feel Safer Than Better Ones

Behavioral economics helps explain some of this through status-quo bias. We have a tendency to prefer existing conditions simply because they already exist. The familiar carries psychological advantages that the unknown does not, which can produce remarkably irrational behavior.

People tolerate inefficient processes because they understand them. Organizations keep ineffective employees because replacing them would be disruptive. Executives avoid difficult conversations because today's dysfunction seems more manageable than tomorrow's uncertainty.

At a much more serious human level, something related can occur when people remain in or return to harmful relationships. That phenomenon cannot be explained simply by status-quo bias—coercive control, attachment, economic dependence, safety concerns, trauma responses, and other factors can be involved.

 This illustrates an important feature of human judgment: the known can sometimes feel safer than an uncertain alternative even when the known situation is demonstrably harmful. Research on people returning to abusive relationships shows just how complex those forces can be.

Organizations do their own version of this. Everyone knows that a particular employee is difficult or that a particular process is dysfunctional. Everyone definitely knows which supervisor avoids accountability.

But everyone has also learned how to navigate around those problems. A new leader disrupts that equilibrium.

The old dysfunction was costly, but predictable. Change introduces uncertainty. And people frequently underestimate how powerful that distinction is.

Hoping for B While Rewarding A

In 1975, Steven Kerr published what became a classic management article: On the Folly of Rewarding A, While Hoping for B.

His central observation was simple and devastating: organizations often say they want one behavior while their reward systems encourage another. The Academy of Management later described the article as a classic because the problem remained so recognizable decades later.

Turnarounds frequently fail for exactly this reason.

Leadership says it wants accountability—B.

But when an employee receives accountability and complains loudly enough, senior leaders pressure the manager to back off.

They have just rewarded A.

Leadership says it wants higher standards—B.

But when enforcing those standards causes turnover, executives criticize the leader for the temporary staffing disruption.

Again, A gets rewarded.

Leadership says it wants culture change—B.

But employees who undermine the new leader use political relationships, complaints, selective information, and organizational history to protect the existing culture, and senior leadership accommodates them.

“A” wins again.

Then executives are surprised when B never arrives.

This is more than inconsistency. It creates an incentive system. Employees learn what actually works. If resistance causes leadership to retreat, resistance becomes rational. If enough complaints can weaken accountability, complaints become a strategy. If political influence protects poor performance, cultivating political influence becomes more valuable than improving performance.

Eventually the organization gets exactly what its incentives have been purchasing.

The Complaint Asymmetry

There is another problem during a turnaround. The people benefiting from improvement are often quieter than the people losing something because of it.

Imagine twenty employees who appreciate clearer expectations, stronger standards, and protection from dysfunctional coworkers. They may simply come to work and do their jobs.

Now imagine three employees who previously enjoyed unusual informal authority, weak accountability, preferred treatment, or the ability to manipulate outcomes. They have lost something. Those three people may generate complaints, accusations, alliances, conversations with executives, and stories about deteriorating morale.

Senior leadership can then make a serious reasoning error: they confuse the volume of resistance with evidence about the quality of the change. Resistance is just information. But resistance does not tell you, by itself, whether the leader is doing something wrong.

Sometimes people resist because leadership is poor or communication is inadequate. Sometimes they resist because a change is genuinely misguided. And sometimes they resist because the change has finally reached the behavior that needed changing.

The responsibility of senior leadership is to distinguish among them. Counting complaints will not accomplish that.

Every Turnaround Has a Transition Cost

This may be one of the most common mistakes executives make when they hire someone to repair a badly functioning operation. They expect the improvement curve to move smoothly upward. Real change often looks messier.

Suppose a department has tolerated poor performance for years. Several employees contribute heavily to the dysfunction. Others have learned to compensate for them. High performers are exhausted. Informal alliances protect some employees from consequences.

Then accountability arrives.

A few employees leave. One is terminated. Several become angry. Staffing temporarily worsens. Complaints increase. Morale scores may fluctuate. Productivity may temporarily decline while new people are recruited and trained. From a distance, executives can look at the disruption and conclude: the new leader is creating problems.

Maybe. But there is another possibility.

They are finally seeing the cost of problems that existed all along. The organization had lived with the dysfunction for so long that the resulting costs no longer looked like warning signs. They looked normal.

Turnaround leadership makes those hidden costs visible.

The Leadership Lottery

There is something economically irrational underneath all of this.

Some organizations approach turnaround leadership almost like purchasing a lottery ticket. They want an enormous return for a very small commitment.

·       Give us radically better culture. But don't upset influential people.

·       Improve accountability. But don't generate complaints.

·       Increase productivity. But don't change staffing too aggressively.

·       Reduce turnover. But don't (you dare) remove the people causing good employees to leave.

·       Change the culture. But preserve its existing power structure.

·       Improve performance. But don't allow performance to dip temporarily while the system is rebuilt.

That is organizational get-rich-quick thinking.

The problem is hardly new. Ancient wisdom warned against the impulse to pursue rapid returns rather than the patient accumulation that produces durable prosperity. Behavioral and decision science now gives us additional language for the same tendency.

When leaders become disproportionately focused on immediate returns, their decision horizon contracts. Short-term KPIs begin carrying more weight than the long-term health of the system producing them. Temporary disruption looks like failure and necessary investment looks like waste. A difficult but corrective staffing decision looks more dangerous than preserving a familiar dysfunction.

Eventually, leaders can start managing the metrics instead of managing the system that produces them.

That distortion can affect staffing, budgeting, marketing, operations, and especially the evaluation of change leaders. A leader rebuilding a damaged system may temporarily produce worse-looking numbers precisely because previously hidden problems are being exposed, weak systems are being replaced, accountability is increasing, or investments are being made whose returns have not yet arrived.

Punishing the leader for that temporary disruption often protects today's KPI at the expense of tomorrow's organization.

The economic principle applies beyond money: large sustainable improvements usually require proportionate investment. The currency may be time, disruption, political capital, temporary performance pressure, difficult conversations, or uncomfortable staffing decisions.

Organizations hoping for transformational returns while refusing meaningful organizational investment are essentially buying leadership lottery tickets.

Occasionally they may get lucky.

Usually they don't.

The Bottleneck Is Revealed

A poor-performing employee can damage a team. A weak supervisor can damage a department. Those are real problems, and the people responsible for them should be held accountable.

But there is a more important leadership question: what allows the problem to persist?

When managers recognize a problem, attempt to address it, and are repeatedly prevented from doing so by the leaders above them, the source of the continuing dysfunction becomes clearer. What looks like an employee problem may actually be evidence of a constraint much higher in the organization.

Peter Drucker famously captured this with a simple metaphor: “The ‘bottleneck’ is, after all, always ‘at the head of the bottle.’”

Organizational change has a way of revealing that bottleneck.

Every level of leadership has a ceiling determined partly by the willingness of the level above it to permit that leader to lead. John Maxwell calls this the leadership lid.

A turnaround manager cannot create sustained accountability if executives repeatedly exempt people from accountability. They cannot rebuild culture if senior leaders protect the informal system producing the old culture.

They cannot build accountability around standards that senior leadership is willing to compromise whenever there is resistance. Change does not move the bottleneck upward. It reveals where the bottleneck was all along.

When Narrative Replaces Evidence

One of the more dangerous moments in organizational conflict occurs when people begin telling stories about performance instead of examining the evidence of performance. This becomes especially consequential during organizational change, when disruption is visible immediately but the benefits of the changes may take longer to appear.

“This leader has destroyed productivity.”

“Morale has never been worse.”

“Everything was better before.”

“Turnover is out of control.”

“The previous system worked better.”

Repeated often enough, and by enough people, claims like these can take on the appearance of established fact. Repetition does not make a false claim true, of course, and neither does the number of people who believe it. A demonstrably false statement remains false regardless of its popularity.

Even the leader being criticized can begin to wonder whether the repeated statements are true.

But something more subtle can happen. Once a claim becomes sufficiently familiar and socially reinforced, people may stop evaluating it as a claim at all. They no longer ask who said it, what evidence supports it, whether the comparison is valid, or what the actual performance measures show. Over time, the story can become accepted as fact, even when no one has stopped to verify whether it is actually true.

That creates a particular danger for senior leaders evaluating a change leader. Instead of using objective performance measures to test the emerging narrative, they may begin using the narrative to interpret the performance measures. Improvement can be discounted, temporary disruption can be exaggerated, and ambiguous data can be interpreted as confirmation that the leader is failing.

This is why major organizational change should begin with a clear baseline of the performance measures that matter. Leaders should know what existed before the intervention, track what happens during it, and, when possible, determine what persists afterward. Without that baseline, organizational memory becomes remarkably vulnerable to revision.

Not every important leadership outcome can be reduced to a KPI, and attempting to do so would create its own distortions. But measurable outcomes provide an important constraint on organizational storytelling because they force perception to contend with evidence.

I recently watched this become extraordinarily useful in a turnaround situation. Before the leader departed, they compiled key performance measures from the period before their arrival, the point when they assumed leadership, and the period immediately before leaving.

The comparison was striking. The narrative circulating inside the organization suggested deterioration, while the actual performance record showed substantial improvement on every measure except one: turnover.

Even that exception required context. A small group of employees was contributing disproportionately to conflict and instability within the department, including behavior that made it difficult to retain new employees who resisted joining the existing informal coalition. The change leader repeatedly attempted to address the underlying conduct through normal accountability processes but was prevented by senior leaders from taking meaningful corrective action.

The result was a particularly destructive form of organizational contradiction: the leader remained accountable for the KPI while being denied authority over one of its primary drivers. Turnover could then be cited as evidence of the leader's failure even though decisions above that leader had materially constrained the ability to correct the conditions contributing to it.

This pattern extends far beyond any one organization or industry. When employees discover that formal authority can be bypassed through political relationships, complaints, informal influence, or appeals to leaders higher in the hierarchy, the accountability system can effectively reverse itself.

In its more destructive form, an employee facing legitimate accountability can redirect scrutiny toward the leader by making accusations of misconduct or mistreatment. The resulting investigation may ultimately find the allegations unsupported, but the process itself can still damage the leader's credibility, create uncertainty among the team, consume enormous amounts of time and attention, and effectively suspend the accountability that triggered the complaint in the first place.

This creates an important organizational vulnerability. Mechanisms intended to protect employees from abusive leadership are essential, but when allegations themselves are treated as evidence, those same mechanisms can be exploited by employees seeking protection from legitimate accountability. The leader can become consumed with defending against accusations while the underlying conduct continues largely untouched.

Once that happens, the change leader faces an almost impossible assignment: improve the outcomes produced by a system while being prohibited from changing some of the people or behaviors producing them.

Senior leadership can then contribute to the very KPI deterioration it later attributes to the change leader. The causal chain is not difficult to see: senior leaders block intervention and discipline; protected behavior continues; good employees leave, disengage, or retreat into self-protection; turnover and other performance measures suffer; and those deteriorating measures are then cited as evidence that the change leader is ineffective.

That is a significant causal problem because the KPI itself may be perfectly accurate. The turnover happened. The error lies in the causal attribution—the explanation assigned to why it happened.

This is a common challenge for middle managers across sectors. They are assigned accountability for outcomes without being granted commensurate authority over the people, systems, and decisions driving those outcomes. When a subordinate leader identifies a constraint, but senior leadership prohibits its removal, the resulting performance problem tells you something about the head of the bottle, not merely the person whose name appears beside the KPI.

This is why evidence matters so much when organizational narratives become politically useful. Data cannot explain every leadership outcome, but it can force confident stories about performance to confront what actually happened.

Human beings are talented storytellers.

Graphs are less politically imaginative.

The Natural Experiment

In the above story, something interesting eventually happened. The turnaround leader left.

People who had argued that this leader was responsible for the problems believed performance would improve after the departure. That created an unintended organizational experiment. There were now two competing explanations.

Explanation One: The turnaround leader had damaged the organization. Remove that leader and performance should improve.

Explanation Two: The turnaround leader had been producing much of the improvement despite internal resistance. Remove that leader and performance should begin drifting toward its previous condition.

Time would tell.

That is one of the advantages of measurable outcomes. Eventually reality gets another vote.

Before You Hire a Change Leader

Executives should therefore ask something more difficult before hiring someone to fix a troubled organization. Asking, “Do we want better results?” is almost useless. Everyone says yes.

Instead ask: What people decisions are we prepared to support? What temporary disruption are we willing to tolerate? Which long-standing practices are genuinely negotiable? How will we respond when influential employees resist? How will we distinguish legitimate complaints from resistance to accountability? What performance measures will tell us whether the change is actually working?

And perhaps the hardest question: What are we willing to let change in order to get the change we say we want? That question reveals far more than the strategic plan.

Organizations eventually receive the behaviors they reward, the standards they enforce, and the dysfunction they protect.

You cannot keep rewarding A and remain surprised that B never shows up.

#Transformationship